Difference between revisions of "Notes20111024"
From Net-SNMP Wiki
| Line 36: | Line 36: | ||
** Wiki captcha is broken - "socket closed". | ** Wiki captcha is broken - "socket closed". | ||
* roundtable | * roundtable | ||
| + | ** hardaker: not much lately | ||
| + | ** dts12: work | ||
| + | ** tanders: 5.7 on Solaris | ||
| + | ** bvanassche: memory leaks | ||
| + | ** tanders: debian to 5.7.1 | ||
Revision as of 20:03, 24 October 2011
Net-SNMP Meeting
2011-10-24
2011-10-24
| Previous: | 2011-09-26 |
| Next: | 2011-11-14 |
This is the meeting agenda/notes for the Meeting held on 2011-10-24.
Date and Time: 2011-10-24 19:00 UTC
- Next Date
- nov 14th
- Releases
- 5.7.1 (should be out)
- what's next?
- Nothing immediate; think of 5.5.x and 5.6.x after December
- Administrative
- Git
- git describe produces strange results
- ie, on 5.5 it shows a tag of v5.4
- hardaker suspects this is because the 5.4 branch is merged into 5.5, and the v5.4 tag was released more recently than any 5.5 tags
- ie, on 5.5 it shows a tag of v5.4
- ChangeLog generation
- hardaker notes that the current ChangeLog generation done by git2cl may not be accurate and would like to consider alternatives
- Discussion:
- what are the goals of the changelog?
- A detailed indication of what has happened to a particular development line
- Raw material for CHANGES/NEWS files
- how useful/used/important do we with think the file is?
- probably not essential
- One of the better documentation introductions!
- users rely on a change log to take upgrade decisions. So I think that a good change log / release notes are really important.
- as long as we meet #1 in some way, is another choice ok?
- Sure
- what are the goals of the changelog?
- git describe produces strange results
- autoconf
- rehashing of previous discussion surrounding use of latest autoconf on all branches
- tanders brings up a point of older distros not being able to update to a later release on a patch tree
- some distros might force-run autoreconf?
- some distros might patch configure but don't have the latest autoconf available and thus if we use later features they're stuck.
- 6.0
- Git
- Technical
- Secondary address support for Linux in IP-MIB is broken.
- ifaliases on linux are broken because of : searching is likely wrong; bvanassche hopes to look at it in a few weeks.
- Wiki captcha is broken - "socket closed".
- Secondary address support for Linux in IP-MIB is broken.
- roundtable
- hardaker: not much lately
- dts12: work
- tanders: 5.7 on Solaris
- bvanassche: memory leaks
- tanders: debian to 5.7.1
